Home > Papers

 
 
Gap-based Belief Selection in Point-based Value Iteration for POMDP
Qi Feng * #,Xuezhong Zhou,Houkuan Huang,Xiaoping Zhang
School of Computer Science and Information Technology, Beijing Jiaotong University
*Correspondence author
#Submitted by
Subject:
Funding: 教育部博士点基金（No.20050004008）
Opened online: 4 February 2009
Accepted by: none
Citation: Qi Feng,Xuezhong Zhou,Houkuan Huang.Gap-based Belief Selection in Point-based Value Iteration for POMDP[OL]. [ 4 February 2009] http://en.paper.edu.cn/en_releasepaper/content/28403
 
 
Partially Observable Markov Decision Processes (POMDPs) provide powerful mathematical models for decision making under uncertainty. Among the algorithms for solving POMDP, point-based value iteration algorithms are effective. In a point-based algorithm, belief selection is a key step. In this paper we propose a belief selection method based on the uncertainty of belief point, which is named Gap-based belief selection. The experimental results indicate that this method is effective to gain an approximate discounted reward using fewer belief states than the PBVI and Distance-based point-based algorithms.
Keywords:POMDP;value iteration;point-based algorithm;belief selection;uncertainty
 
 
 

For this paper
  • ● PDF (0B)
  • ● Revision 0 　　
  • ● Print this paper
  • ● Recommend this paper to a friend
  • ● Add to my favorite list

    Saved Papers

    Please enter a name for this paper to be shown in your personalized Saved Papers list

Tags
Add yours

Related Papers

Statistics
PDF Downloaded 215
Bookmarked 0
Recommend 5
Comments Array
Submit your papers